Neural correlates of recalled sadness, joy, and fear states: a source reconstruction EEG study

TL;DR

This study uses EEG to identify brain activity patterns linked to recalled emotions like sadness, fear, and joy, which could help people who cannot express their feelings.

Contribution

The study demonstrates the feasibility of detecting and classifying emotional states using non-invasive EEG source reconstruction.

Findings

Common and distinct brain regions were identified for sadness, fear, and joy during emotional recall.

Right temporal and left superior frontal areas were more active during sadness, left limbic for fear, and right orbitofrontal for joy.

The findings suggest potential for using EEG in brain-computer interfaces to identify emotional states in locked-in syndrome patients.

Abstract

The capacity to understand the others’ emotional states, particularly if negative (e.g. sadness or fear), underpins the empathic and social brain. Patients who cannot express their emotional states experience social isolation and loneliness, exacerbating distress. We investigated the feasibility of detecting non-invasive scalp-recorded electrophysiological signals that correspond to recalled emotional states of sadness, fear, and joy for potential classification.

The neural activation patterns of 20 healthy and right-handed participants were studied using an electrophysiological technique. Analyses were focused on the N400 component of Event-related potentials (ERPs) recorded during silent recall of subjective emotional states; Standardized weighted Low-resolution Electro-magnetic Tomography (swLORETA) was employed for source reconstruction. The study classified individual patterns of brain activation linked to the recollection of three distinct emotional states into seven regions of interest (ROIs).

Statistical analysis (ANOVA) of the individual magnitude values revealed the existence of a common emotional circuit, as well as distinct brain areas that were specifically active during recalled sad, happy and fearful states. In particular, the right temporal and left superior frontal areas were more active for sadness, the left limbic region for fear, and the right orbitofrontal cortex for happy affective states.

In conclusion, this study successfully demonstrated the feasibility of detecting scalp-recorded electrophysiological signals corresponding to internal and subjective affective states. These findings contribute to our understanding of the emotional brain, and have potential applications for future BCI classification and identification of emotional states in LIS patients who may be unable to express their emotions, thus helping to alleviate social isolation and sense of loneliness.

Figures